And she's off! Hilary Clinton announces her intentions to run.

[Meshakhad]

I do not like Hillary, but I will probably end up voting for her.

Simply put, her positions seem very close to mine. My problem is with her character.

[mythbuster43]

Let the poo flinging commence.

Stay classy Donald.

... so how satisfied are the people you answer to, Mr. chief executive?

At least her husband didn't leave her for some younger model, Mr. Trump.

[Lt. Fred]

What sort of working political system could you devise that would elect an executive?

Can you explain that a bit?

Something else just occurred to me. The whole Hillary '16 thing reminded me of something, some sort of feeling like I've seen this before. Where she was the presumptive nominee, that the party's nomination was hers to lose. Why does that sound familiar?

The US system has a separately elected executive, which, I think, is a good move (in my political system, legislature and executive are the same, which is silly). But the US voting system is silly. Could there be one that isn't?

What aspects of it do you find objectionable? The electoral college? That it's first-past-the-post?

It's a two party system. Now, that's very easy to remedy at the Congressional level - simply close down the Senate, and elect the House through state-wide proportional representation; the entire state becomes an electorate, and you elect a number of congresspeople corresponding to state population. If the Yellow Party gets 8% of the vote, they get 8% of the congresspeople instead of, in the current system, 0%. Do that and you have a decent voting system.

But how do you develop a non-idiotic voting system when ultimately you elect only one person?

I find the system you propose here idiotic because how do you get rid of a bad lawmaker who has large amounts of influence within his or her chosen party? If there are, say, twenty-five seats to be allocated, and the Yellow Party gets two of them, how can the general electorate--or even those voters who voted for the Yellow Party--ensure that those two seats to go people who aren't bad?

Ah yes. Well this is why you have extremely strong political parties. If individuals don't vote with the party always, they party kicks them out.

You can also devise systems that allow you to do this. You can vote "below the line". So typically, a voter card will look like this:



Most people will vote above the line - 1, 2, 3, ect. Their vote will go to the first party they vote for until that party loses, at which point it will flow to the next one and the next one.

If the party nominates some dingbat, you can vote below the line for individual candidates. So you vote 1 for their second candidate, etc. And you end up putting their dingbat candidate well down the list. This means you'll be writing numbers a lot longer.

That said, this is extremely rare in systems with strong political parties. You almost always get exactly the same sorts of candidates - even if you don't, the candidate has little influence within the party, so the candidate matters little. Only policy matters. The system where you vote for what you want government to do is called democracy. The alternative system - where you vote for a person to do whatever that person wants - is something else.

Also, how can independent candidates get elected in this system?

LOADS more independent candidates get elected in this system (depending on how you define independent). In the current system 100% of real power is in the hands of the Democrats and the Republicans. You either vote to get rid of one or the other, and the party you incidentally vote for decides what to do with the power you've incidentally given them (you have no control, because who else are you going to vote for?). So independents don't matter. In Australia's weak PR system in our senate, we've always had multiple independent candidates, who tend to hold the balance of power.

In this system, most governments will be coalition governments. That means parties that are neither Democratic nor Republican will form government. "Independent" members - that is, third parties - will dominate. You totally control what the do with the power you give them, because if they don't do what you voted for them to do, you can vote for fifteen other parties. So you control their policy agenda.

And further, if this is being done state-by-state, this system still has the issue of reducing to an "idiotic" system in a state with only one House seat (of which there are currently seven), and nearly so in states with very few representatives.

That is indeed a problem. It won't work if you only have a small population. Maybe increase the size of the House, after repealing the senate?

[dpareja]

1. Ah, you're talking about a Single Transferable Vote system (if I understand you correctly), rather than a party list system (either fully as in Israel or partially as in Germany or Japan). The issue I have with single transferable vote is that its one-seat reduction, Hare or instant runoff, is absurd: http://zesty.ca/voting/sim/

2. I agree, in STV independents can and do get elected. As I said, I thought you were talking about a list system.

3. I find it a bit interesting that you advocate abolishing the US Senate, when the Australian Senate gives 12 seats to both New South Wales and Tasmania, the former having a population about 14 times that of the latter. Further, the Australian Capital Territory, having a population more than two-thirds that of Tasmania, receives only 2 seats. Do you also advocate abolishing the Australian Senate?

4. The US House seats are apportioned by the Huntington-Hill method with the number of seats fixed at 435. To give every state two seats under that method, there would have to be, using the populations from the 2010 Census, at least 773 seats, and to give every state three seats (which I believe is the minimum constituency size in the Republic of Ireland), at least 1342 seats. Even the former is larger than the EU Parliament, which has 751 members.

[davedan]

The US electoral college system is a fairly silly way to elect anyone. Particularly when it is a vote for President and Vice-President. A direct vote for president would seem to be less silly.

[Lt. Fred]

1. Ah, you're talking about a Single Transferable Vote system (if I understand you correctly), rather than a party list system (either fully as in Israel or partially as in Germany or Japan). The issue I have with single transferable vote is that its one-seat reduction, Hare or instant runoff, is absurd: http://zesty.ca/voting/sim/

You never have runoffs in a Hare-Clarke system, unless it's for a single person (ie, a President). I agree that's bad.

3. I find it a bit interesting that you advocate abolishing the US Senate, when the Australian Senate gives 12 seats to both New South Wales and Tasmania, the former having a population about 14 times that of the latter. Further, the Australian Capital Territory, having a population more than two-thirds that of Tasmania, receives only 2 seats. Do you also advocate abolishing the Australian Senate?

Absolutely, assuming we introduce a STV or some other PR system in the lower house the day before. The idea that people's interests are largely dependent on which state they live in is obviously silly.

4. The US House seats are apportioned by the Huntington-Hill method with the number of seats fixed at 435. To give every state two seats under that method, there would have to be, using the populations from the 2010 Census, at least 773 seats, and to give every state three seats (which I believe is the minimum constituency size in the Republic of Ireland), at least 1342 seats. Even the former is larger than the EU Parliament, which has 751 members.

Another way to do it would be to combine small states together.

[dpareja]

1. Ah, you're talking about a Single Transferable Vote system (if I understand you correctly), rather than a party list system (either fully as in Israel or partially as in Germany or Japan). The issue I have with single transferable vote is that its one-seat reduction, Hare or instant runoff, is absurd: http://zesty.ca/voting/sim/

You never have runoffs in a Hare-Clarke system, unless it's for a single person (ie, a President). I agree that's bad.

But you do: every vote reallocation after the initial first-choice votes are counted is a runoff of sorts, and once only one seat is left to be awarded the system is Hare, even if it's using lower choices from many ballots, many only counting fractionally.

4. The US House seats are apportioned by the Huntington-Hill method with the number of seats fixed at 435. To give every state two seats under that method, there would have to be, using the populations from the 2010 Census, at least 773 seats, and to give every state three seats (which I believe is the minimum constituency size in the Republic of Ireland), at least 1342 seats. Even the former is larger than the EU Parliament, which has 751 members.

Another way to do it would be to combine small states together.

More like merge small states into larger neighbouring ones (although you can just make one huge (in area) state out of Montana, Wyoming, North Dakota and South Dakota), and that doesn't solve Alaska (1 seat), or if we want, at minimum, three-seat constituencies, Hawaii (2 seats).

[Lt. Fred]

1. Ah, you're talking about a Single Transferable Vote system (if I understand you correctly), rather than a party list system (either fully as in Israel or partially as in Germany or Japan). The issue I have with single transferable vote is that its one-seat reduction, Hare or instant runoff, is absurd: http://zesty.ca/voting/sim/

You never have runoffs in a Hare-Clarke system, unless it's for a single person (ie, a President). I agree that's bad.

But you do: every vote reallocation after the initial first-choice votes are counted is a runoff of sorts, and once only one seat is left to be awarded the system is Hare, even if it's using lower choices from many ballots, many only counting fractionally.

Oh right. I thought you meant a runoff election.

I don't see the problem with allocating preferences.

4. The US House seats are apportioned by the Huntington-Hill method with the number of seats fixed at 435. To give every state two seats under that method, there would have to be, using the populations from the 2010 Census, at least 773 seats, and to give every state three seats (which I believe is the minimum constituency size in the Republic of Ireland), at least 1342 seats. Even the former is larger than the EU Parliament, which has 751 members.

Another way to do it would be to combine small states together.

More like merge small states into larger neighbouring ones (although you can just make one huge (in area) state out of Montana, Wyoming, North Dakota and South Dakota), and that doesn't solve Alaska (1 seat), or if we want, at minimum, three-seat constituencies, Hawaii (2 seats).
[/quote]

Yeah, that'd do it. Probably Hawaii can stay a state, but contribute to California or something.

[dpareja]

1. Ah, you're talking about a Single Transferable Vote system (if I understand you correctly), rather than a party list system (either fully as in Israel or partially as in Germany or Japan). The issue I have with single transferable vote is that its one-seat reduction, Hare or instant runoff, is absurd: http://zesty.ca/voting/sim/

You never have runoffs in a Hare-Clarke system, unless it's for a single person (ie, a President). I agree that's bad.

But you do: every vote reallocation after the initial first-choice votes are counted is a runoff of sorts, and once only one seat is left to be awarded the system is Hare, even if it's using lower choices from many ballots, many only counting fractionally.

Oh right. I thought you meant a runoff election.

I don't see the problem with allocating preferences.

Look at the statistical models in the page to which I linked: Hare produces crazy results. STV is just an n-seat generalization of Hare. There's no problem with considering preferences (and indeed preferences should be considered, in my view), but instant runoff is nuts.

[Lt. Fred]

1. Ah, you're talking about a Single Transferable Vote system (if I understand you correctly), rather than a party list system (either fully as in Israel or partially as in Germany or Japan). The issue I have with single transferable vote is that its one-seat reduction, Hare or instant runoff, is absurd: http://zesty.ca/voting/sim/

You never have runoffs in a Hare-Clarke system, unless it's for a single person (ie, a President). I agree that's bad.

But you do: every vote reallocation after the initial first-choice votes are counted is a runoff of sorts, and once only one seat is left to be awarded the system is Hare, even if it's using lower choices from many ballots, many only counting fractionally.

Oh right. I thought you meant a runoff election.

I don't see the problem with allocating preferences.

Look at the statistical models in the page to which I linked: Hare produces crazy results. STV is just an n-seat generalization of Hare. There's no problem with considering preferences (and indeed preferences should be considered, in my view), but instant runoff is nuts.

I think FPTP produces crazy results. Hare-Clarke produces representative results.

[dpareja]

1. Ah, you're talking about a Single Transferable Vote system (if I understand you correctly), rather than a party list system (either fully as in Israel or partially as in Germany or Japan). The issue I have with single transferable vote is that its one-seat reduction, Hare or instant runoff, is absurd: http://zesty.ca/voting/sim/

You never have runoffs in a Hare-Clarke system, unless it's for a single person (ie, a President). I agree that's bad.

But you do: every vote reallocation after the initial first-choice votes are counted is a runoff of sorts, and once only one seat is left to be awarded the system is Hare, even if it's using lower choices from many ballots, many only counting fractionally.

Oh right. I thought you meant a runoff election.

I don't see the problem with allocating preferences.

Look at the statistical models in the page to which I linked: Hare produces crazy results. STV is just an n-seat generalization of Hare. There's no problem with considering preferences (and indeed preferences should be considered, in my view), but instant runoff is nuts.

I think FPTP produces crazy results. Hare-Clarke produces representative results.

I agree that FPTP produces crazy results. It's an awful system only used because of its simplicity of explanation and ease of implementation.

I fail to see how you can look at a system that produces disjoint winning regions for candidates, or where a shift of preference toward a candidate can cause that candidate to lose, and declare its results "representative".